STORAGE SYSTEM COMPRISING FLASH MEMORY, AND STORAGE CONTROL METHOD

• US 20120317333A1
• Filed: 06/07/2011
• Published: 12/13/2012
• Est. Priority Date: 06/07/2011
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Yamamoto, Akira, Yamamoto, Masayuki
• Granted Patent: US 8,862,805 B2
• Time in Patent Office: Days
• US Class Current: 711/103
• CPC Class Codes: G06F 2212/1044 Space efficiency improvement
• CPC Class Codes: G06F 2212/401 Compressed data
• CPC Class Codes: G06F 2212/7208 Multiple device management, e.g. distributing data over multiple flash devices
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0665 at area level, e.g. provisioning of virtual or logical volumes
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays

Abstract

A storage system has a plurality of flash packages, and a storage controller for receiving a write request from a host and sending a write-data write request based on data conforming to this write request to a write-destination flash package. A virtual capacity, which is larger than the physical capacity of the flash package, is defined in the storage controller. The storage system compresses the write data, and writes the compressed write data to the write-destination flash chip.

Claims

1. A storage system, which comprises a plurality of flash packages having a plurality of flash chips having a plurality of blocks, which are data delete units, and a storage controller for receiving a first write request from a host and sending a second write request for write data based on data conforming to this first write request to a write-destination flash package of the plurality of flash packages,

the storage system further comprising:

a first capacity virtualization part for defining in the storage controller a virtual capacity, which is larger than a total value of a physical capacity of the plurality of flash chips of the flash packages; and

a compression part for compressing the write data, wherein the compressed data, which is compressed write data, is written to a write-destination flash chip in the write-destination flash package.

2. A storage system according to claim 1, further comprising:

a data storage status management part for managing the data storage status of the flash package; and

a virtual capacity change part for changing the virtual capacity of this flash package in accordance with the data storage status of the flash package.

3. A storage system according to claim 2, wherein

the flash package data storage status is a compression ratio of this flash package, and

the flash package compression ratio is a value based on the total capacity of pre-compression write data with respect to valid compressed data stored in this flash package, and the total capacity of the valid compressed data stored in this flash package.

4. A storage system according to claim 3, wherein the virtual capacity change part, in a case where the flash package compression ratio has increased, increases the virtual capacity of this flash package, and in a case where this flash package compression ratio has decreased, reduces the virtual capacity of this flash package.

5. A storage system according to claim 4, wherein the virtual capacity change part, in a case where the flash package compression ratio has increased more than a reference value, which is a value based on the virtual capacity of this flash package and the real capacity of this flash package, increases the virtual capacity of this flash package, and in a case where this flash package compression ratio has decreased more than the reference value, reduces the virtual capacity of this flash package.

6. A storage system according to claim 5 further comprising:

a second capacity virtualization part for allocating an empty real page from among a plurality of real pages to a write-destination virtual page, which conforms to the first write-request, of a plurality of virtual pages configuring a logical volume provided to the host, wherein

write data conforming to the second write request is data, which conforms to the first write request and is based on data written to the allocated real page,

in a case where the flash package virtual capacity increases, one or more empty real pages are added to a plurality of real pages based on this flash package in proportion to the increase in this virtual capacity, and

in a case where the flash package virtual capacity decreases, one or more empty real pages are subtracted from the plurality of real pages based on this flash package in proportion to the decrease in this virtual capacity.

7. A storage system according to claim 3, wherein

the block is a real block,

the first capacity virtualization part allocates an allocatable real block from among a plurality of real blocks to a compressed data write-destination virtual block,

compressed data is written to the allocated real block,

each of the virtual blocks is configured from a plurality of virtual segments,

the data storage status management part manages an amount of valid compressed data written to a virtual segment and an amount of pre-compression data with respect to this valid data,

a total capacity of the pre-compression write data with respect to the valid compressed data stored in the flash package is a value computed on the basis of the amount of pre-compression data with respect to the written valid data for each virtual segment provided by this flash package, and

a total capacity of the valid compressed data stored in the flash package is a value computed on the basis of an amount of valid compressed data stored in this flash package for each virtual segment provided by this flash package.

8. A storage system according to claim 2, wherein

the flash package comprises the data storage status management part and a status notification part for notifying the storage controller of the data storage status, and

the storage controller comprises the virtual capacity change part.

9. A storage system according to claim 2, wherein the flash package comprises the data storage status management part and the virtual capacity change part.

10. A storage system according to claim 2, comprising:

a second capacity virtualization part for allocating an empty real page from among a plurality of real pages to a write-destination virtual page, which conforms to the first write-request, of a plurality of virtual pages configuring a logical volume provided to the host, wherein

write data conforming to the second write request is data, which conforms to the first write request and is based on data written to the allocated real page.

11. A storage system according to claim 10, wherein

in a case where this flash package virtual capacity increases, one or more real pages based on this flash package are added in accordance with the extent of the increase in this virtual capacity, and

in a case where this flash package virtual capacity decreases, one or more real pages based on this flash package are subtracted in accordance with the extent of the decrease in this virtual capacity.

12. A storage system according to claim 10, comprising:

a transfer decision part for deciding, based on the flash package data storage status, that data is to be transferred, between the flash packages, from a transfer-source real page to a transfer-destination real page; and

a transfer part for transferring data, between the flash packages, from the transfer-source real page to the transfer-destination real page in accordance with the decision of the transfer decision part.

13. A storage system according to claim 12, wherein management information associated with the transfer-destination real page is updated on the basis of management information associated with the transfer-source real page.

14. A storage system according to claim 1, wherein the flash package performs a delete process to a real block, which is a delete unit of the flash chip, and prior to executing this delete process, selects valid data for the storage controller from among the compressed data stored inside the delete process-targeted real block, allocates a new real block, and writes to the allocated real block the selected valid data in the compressed state.

15. A storage system according to claim 1, further comprising a block release part for releasing a real block, in which a specific pattern is repetitively written, from a virtual block.

16. A storage system according to claim 15, wherein

the storage controller notifies the flash package of the write destination of the specific pattern, and

the flash package comprises the block release part.

17. A storage control method for a storage system, which has a plurality of flash packages having a plurality of flash chips, which has a plurality of blocks, which are data delete units, and a storage controller for receiving a first write request from a host and sending a second write request for write data based on data conforming to this first write request to a write-destination flash package of the plurality of flash packages,

the storage control method comprising:

defining in the storage controller a virtual capacity, which is larger than a total value of a physical capacity of the plurality of flash chips of the flash packages;

compressing write data; and

writing compressed data, which is compressed write data, to a write-destination flash chip in the write-destination flash package.